question about ODDRATIO

[litun-fkby]

Hello,
i ues the vardict 1.8.3 to call snv/indel and i'm confused about the ODDRATIO
there has one snv record as follow:
chr6 117885637 . A T 116 PASS SAMPLE=2-M10_lot20230614;TYPE=SNV;DP=1469;VD=28;AF=0.0191;BIAS=2:2;REFBIAS=714:695;VARBIAS=2:26;PMEAN=35.4;PSTD=1;QUAL=24.3;QSTD=1;SBF=0;ODDRATIO=13.34354;MQ=60;SN=1.8;HIAF=0.0130;ADJAF=0;SHIFT3=0;MSI=4;MSILEN=1;NM=1.5;HICNT=18;HICOV=1380;LSEQ=GATGTCTAGCTAAAACTTAA;RSEQ=ACGAATTCTGATCATCTATA;DUPRATE=0;SPLITREAD=0;SPANPAIR=0 GT:DP:VD:AD:AF:RD:ALD 0/1:1469:28:1409,28:0.0191:714,695:2,26

and i calculate the ODDRATIO:
oddratio=（26714)/(2695)=13.3553

so,my questions is:
1.the value is close to the record but different, is there any correction here?
2.if the value 0 in REFBIAS or VARBIAS, or ODDRATIO < 1 , is there any correction here?
3.the ODDRATIO=13.34354, and the SBF=0 ，it seem like the strand reads distribution support the ref and variant is different，but the BIAS=2:2 which means no strand bias in ref an variant reads, if i want to ues these info to filter variant, is there some advices orrecommended value ?
4.i notice that other tool to call snv have the info SOR (such as GATK ),which value range is 0-1,what's the difference?
https://gatk.broadinstitute.org/hc/en-us/articles/360046786792-StrandOddsRatio

Hope your reply.
Thanks!